HEALTH SERVICES FOR SEXUALLY ASSAULTED. COMPLAINANTS’ AND JUDICIAL USE OF A SELF-REFERRAL CENTRE; CASE-FLOW FROM ASSAULT TO LEGAL OUTCOME

(1)

HEALTH SERVICES FOR SEXUALLY ASSAULTED.

COMPLAINANTS’ AND JUDICIAL USE OF A SELF-REFERRAL CENTRE;

CASE-FLOW FROM ASSAULT TO LEGAL OUTCOME

Helle Margrethe Floor Nesvold

Emergency Medical Agency, City of Oslo Institute of Forensic Medicine, Oslo University

Institute of Psychiatry, Section of adult psychiatry, Oslo University Hospital, Ulleval

(2)

© Helle Margrethe Floor Nesvold, 2010

Series of dissertations submitted to the Faculty of Medicine, University of Oslo No. 958

ISBN 978-82-8072-584-4

reproduced or transmitted, in any form or by any means, without permission.

Cover: Inger Sandved Anfinsen.

Printed in Norway: AiT e-dit AS.

Produced in co-operation with Unipub.

The thesis is produced by Unipub merely in connection with the

thesis defence. Kindly direct all inquiries regarding the thesis to the copyright

(3)

3

ACKNOWLEDGEMENTS

This project has grown slowly, like the trees on our Northern coastline.

Years of clinical experience as a general practitioner/medical forensic examiner, and many discussions on the premises for our Sexual Assault Centre (SAC), had evoked curiosity on many topics.

Themes had to be sorted out and research methodology had to be learned: the drafting of a project, writing applications for funding and ethical approval, how to make registration forms and deal with dilemmas in the registration, how to deal with statistics and statistical data programs, how to write a manuscript. Plunging into a detailed registration permitted the exploration of a continent familiar from practical work, whereas cartography and structure developed by and by.

After some initial sidesteps on topics of forensic interest; e.g. injury patterns and toxicological results; I settled on focusing the use of SAC services, case-flow from assault to SAC/police and attrition of cases and SAC information.

In retrospect; starting with a multi-centre study was rather a risk sport for a beginner but evoked questions regarding interpretation of attending patterns and barriers against attending.

Further exploration of attending patterns led to reflections on the paradoxes related to forensic medical examination (FME) in a SAC setting. This examination is meticulous and very expensive for the centres if these are to provide work of quality i.e. in competence and sufficient resources/practical facilities, and is performed on behalf of the police and jurisprudence. Yet, our juridical authorities do not engage in defining any standards of quality, neither in providing education nor framework for clinical forensic medical work in general. Considering our small and scattered population, the few forensic institutions and the many small SACs, we urgently need a common organisation for forensic services.

Today, the costs of forensic casework at SACs are pushed on to the local municipalities, with a minor contribution from the local police. A set of guidelines including forensic instructions have been developed by health authorities without any system for up-dating according to forensic development. As for the rehabilitation of victims, which definitely should be of concern for health authorities, no minimum standards or guidelines are defined – although some recommendations have been presented by juridical authorities. And the politicians seem to believe that all problems are solved by saying ”let there be SACs”.

Through this project I wanted to show the benefits of self-referral centres compared to traditional police-requested examinations at forensic institutes, that the police gain a lot thereby without carrying the extra costs. There is also a considerable attrition with

consequences for legal outcome since the police do not utilise all available medical evidence and victims back off. Last, but not least, victims’ medical needs obviously deserve more attention, in practical medicine as well as in research.

Hopefully, this project may contribute to further improvements.

However, without encouragement and support, this project might have floundered:

Berit Schei encouraged me to begin and introduced me to Norvold, a Nordic research network funded by the Nordic Council of Ministers. This network initiated the comparative study and introduced me to several clever women that I am proud to know, amongst these my cowriters in the comparative study; Guðrún Agnarsdóttir, Anne-Marie Worm and Ursula Vala.

Ole Christian Hjemdal gripped me by my neck when I thought I’d better quit the project before starting the main work, and offered a guest link to the NKVTS (National Centre for Violence and Traumatic Stress Studies), a multidisciplinary research centre.

(5)

5 Meeting Svein Friis, my one tutor, was like attending a master class whilst still playing the piano with one finger. A quiet question from him could turn my thoughts in new directions, new perspectives unfolding. He just asked and patiently left me to find the answers, then later advising on the presentations.

Kari Ormstad, my other tutor, steadily kept me going by her generous support and sense for language, her indescribable humour and an inspiring stamina.

Characteristic for both tutors, they are hard working and honest, dedicated and blessedly devoid of pointed elbows.

At the Oslo Police District, Veslemøy Grytdahl assisted in combing the police registers and archives.

At the Forensic Institute, Oslo University Hospital, Eva Stålstrøm collected the relevant results from trace evidence analysis.

Norwegian Women’s Public Health Association and EXTRA funds from the Norwegian Foundation for Health and Rehabilitation supplied funding, and were understanding when I had to postpone the main start of the project. Actually, being recommended by the Norwegian Women’s Public Health Association is an honour, as this organisation has addressed many health and social issues in the Norwegian society; women indeed are capable.

Oslo University Hospital, Ulleval, friendly supplied a variety of practical support. Division of psychiatry provided affiliation for the project. Division for Research and Development administered the funding. Oddmar Moen gave lessons in SPSS and syntaxes. Leiv Sandvik supervised the statistics and Heidi Thorstensen gave advice on data security. Librarian Irene Refsland and her colleagues taught me to master reference tools and manuscript centrals.

Oslo Emergency Medical Agency, Legevakten, supplied a flexible attachment with the SAC during the research period, supported by the Norwegian Directorate of Health as a basic national competence building programme was requested.

Prior to the research period, the medical work at SAC had been a quite lonely one, the introduction of colleague Henriette M Waitz made a positive difference. Since 2007, the medical staff has expanded further, and at present, SAC is in a new phase of development.

The social service counsellors have always been there, supportive and interactive since SAC’s beginning.

Patients have taught me a lot. I have been struck by how they may suffer, their courage to fight for restitution, and their personal dignity although they may not perceive dignity themselves. Sexual assaults do represent serious health risks, casting long shadows.

The court trials have taught me to appreciate the meticulousness of forensic work and to realise the challenge of being a forensic medical examiner in a care-giving organisation.

Finally, I would like to thank my family and friends who have accepted my years of intensive activity with research project, practical medical work and national implementations. My husband has faithfully handled 90% of the cooking and the children’s’ activities. Our children have grown out of childhood during these years – I hope I have not failed them by spending all this time on work. My old MD father, who used to worry about the enterprise, may soon settle his worries. My resentment to resign, I believe to be a maternal heritage added a dash of paternal stubbornness.

My friends have accepted my absence – I shall re-socialise before long.

I am grateful to you all - most of all to my family and tutors.

(6)

ABBREVIATIONS

Dk: Denmark Fi: Finland

FME: Forensic medical examination FMI: Forensic medical institute Is: Iceland

No: Norway SA: sexual assault

SAC: Self-referral Sexual Assault Centre

(7)

7

PUBLICATIONS INCLUDED IN THE THESIS

The thesis is based on these four papers which are referred to in the text by their roman numerals:

I. Different Nordic facilities for victims of sexual assault: a comparative study Nesvold H, Worm AM, Vala U, Agnarsdottir G.

Acta Obstet Gynecol Scand 2005;84:177 – 83

II. Sexual assault centres: Attendance rates, and differences between early and late presenting cases

Nesvold H, Friis S, Ormstad K.

Acta Obstet Gynecol Scand 2008;87:707 – 15

III. Sexual assault centres and police-reporting - an important arena for medical/legal interaction

Nesvold H, Ormstad K, Friis S.

Ref Type: Submitted to Journal of Forensic Sciences

IV. To be used or not to be, that is the question.

Legal use of forensic and clinical information collected in a self-referral Sexual assault centre

Nesvold H, Ormstad K, Friis S.

Ref Type: Submitted to Journal of Forensic Sciences

(8)

(9)

9

INTRODUCTION

During the last few decades sexual assaults (SA) have been increasingly focused in the media and in professional literature. The main areas of concern have been prevalence, consequences, cultural attitudes, insufficient supportive services, low reporting and conviction rates (1-10).

It is generally acknowledged that sexual assaults represent major social, health and legal problems, yet often embedded in myths and neglected. In a global perspective, sexual assaults contribute to the general oppression of women and children, but also affect males, occur commonly in torture, and wartime rapes are documented since ancient times (11); the latter now defined as a war crime and of much concern for international organisations (12;13).

Although oppression is less prominent in the Western world, treatment of victims is still suboptimal. Legislation and services should be further improved, and developed states governed by law have a particular responsibility to promote improvements and unmask the seriousness of these crimes.

International UN conventions like UNCP (Covenant on Civil and Political Rights), CEDAW (Convention to Eliminate Discrimination against Women) and the CJSV(Convention on Justice and Support for Victims of Crime and Abuse of Power) impose national obligations to address victims’ legal situation/legal rights and the need for preventive measures (14-16). The European Convention of Human Rights ensuring the rights of the accused is also applicable to victims’ rights (7;17-19). WHO has published Guidelines for medico-legal care for victims of sexual violence (1;20;21). Several countries have revised their legislation regarding sexual assault (17;22) and reporting rates increase in Northern Europe (17;22-26) – but not conviction rates (7;17;22), and access to assistance and treatment is highly variable.

Supportive services for victims may be non-existing or self-supportive at women’s shelters.

Access to professional help, counselling and medical treatment may sometimes be linked to examinations at forensic medical institutes (FMI) but is more often scattered for victims themselves to search for. However, an increasing number of self-referral centres are established, as recommended (20;27;28). Victims can approach these centres directly for multidisciplinary services like medical care, psychosocial support and standardised forensic medical examination (FME); but the centres’ organisational fundament vary.

(10)

This diversity of services reflects that medical services originally were requested for legal purposes. During recent times, health and welfare systems gradually have taken on responsibility for rehabilitation. Modern self-referral sexual assault centres (SAC) combine two intentions by providing medicolegal documentation as well as socio-medical assistance.

How purposeful are these services?

There is a need for studies that compare and evaluate results from different services and centres,e.g. self-referral SACs and police-dependent FMIs, their attendance rates and client/case characteristics. So far,such inter-centrecomparisons are rare (29).

There is also sparse knowledge of SACs’ success in reaching the total target group as most SAC-based studies report only on cases which have been attended to.

Attendance rate (number of visitors vs. population served) and incidence of sexual assault are informative of the thresholds for help-seeking. Attendance rates enable inter-centre

comparisons and intra-centre monitoring; differences may indicate threshold variations. Few studies focus the services utilised by those consulting and even less attention has been given those arriving too late for FME. Attending patterns and use of services at self-referral centres reflect victims’ preferences regarding services, i.e. information useful for adapting the services to victims’ needs.

We therefore need studies of early and late attendance and estimates of the fraction of actually assaulted persons that consult the SACs.

Medical contribution to the legal process is also insufficiently explored. Both SACs and FMIs have presented examination results vs. legal outcome (8;30-40), but when other key evidence like e.g. witness testimonies, seldom are included (38;39), interpretation is difficult. SAC- based studies may compare police-reported cases to non-reported cases (41;42); reliable identification of actually police-registered cases evidently being mandatory. Few address the judiciary benefit of performing FME prior to police reporting (43). Case logistics from alleged assault to SAC and further to police has scarcely been touched, neither has utilisation of medical documentation in police investigation. Attrition in rape cases within the legal system has been a topic in criminology, less so the attrition from assault to SAC and from SAC to police. Attrition and legal proceedings in rape cases are also widely discussed in public, and moreknowledge about legal use of SAC services ought to be included in the

(11)

11 discussion. Such information is also relevant for SACs defending the costs of forensic work irrespective of police involvement, as forensic work is expensive.

The present study addresses the challenges mentioned above.

The main aims of the study are:

1. To compare attending rates and case profiles regarding police-reported sexual assaults seen at different medical services: FMIs with police-regulated admission and self- referral SACs

2. To estimate the fraction of actually occurring assaults presented to SAC, and explore differences in case panorama and use of services among early and late attendees 3. To identify SAC cases registered with the police, case logistics from assault to SAC

and police and thereby evaluate:

a) Reporting practices in cases presented in time for FME and those presented later b) Costs and benefits of self-referral FME performed irrespective of police

involvement, i.e. fraction of performed FMEs which is not further utilised due to non-reporting, and police gain in cases of delayed report (FMEs performed 2 days before reporting)

c) Whether police involvement can be sufficiently predicted for selective performance of FME

d) The attrition; percentage of SAC cases reported, and fractions lost to the legal system as not reported, complaint withdrawn or information not collected; as well as reported rapes where the victims did not attend SAC

4. To identify to what extent the police use information collected by the SAC, and thereby evaluate:

a) Predictors for police use of SAC-based information and material; in cases where the information is available for legal use and explored separately for cases with available FME data and cases without

b) Potential contribution of SAC-based casework in the legal process

The results may contribute to improve the services offered, to facilitate SAC/police/victim interaction, and to improve legal use of SAC casework.

(12)

MATERIAL AND METHODS

The project emanates from the Nordic countries, the main SAC in the project being located in Oslo, Norway. The Nordic countries have had differently organised services for sexually assaulted, and they are culturally sufficiently close for comparisons. Regarding Oslo; the city is served by this single SAC and constitutes a single police precinct. The population has been studied for prevalence and incidence of sexual assault. These factors allow a comparison of actually occurring assaults to those seen at the SAC. All SAC cases registered with the Norwegian police can be identified in reliable registers. The given permission to review the police files allowed investigation of case logistics, attrition and police’s use of SAC work.

The present series are descriptive, based on retrospectively collected data from the Oslo Sexual Assault Centre and corresponding police files in police-reported cases. The first section where SACs are compared to forensic medical institutes (FMIs) also includes data from SAC Reykjavik (Iceland), FMI Helsinki (Finland) and FMI Copenhagen (Denmark). Data from Oslo SAC are included in all sections of the project. This SAC was established in 1986 and serves a population of about 500,000 inhabitants, catering to victims 14 yrs old of both genders. The centre is located in the main outpatient emergency ward and defined as a health service. Victims are protected by health care confidentiality unless giving an informed consent to release information, e.g. to the police.

The centre was originally intended for acute cases, but the strict time limits were abandoned as expansion within existing capacity disclosed a variety of needs also among later attendees.

All victims are offered medical care, psychosocial support and a 3 – 4 months’ follow-up at SAC, as well as referrals to other services when appropriate. Patients arriving in time are offered a standardised forensic medical examination, irrespective of police involvement. In unreported cases, trace evidence samples are stored for minimum three months. Victims may freely choose which services to use, all free of charge. SAC’s costs are carried by the health institution housing SAC supported by The National Health Insurance (medical help/treatment) and the police (a modest fee when medical information is requested).

In the mid-nineties, self-referral services for sexually assaulted (SAC) had been organised in two Nordic capitals (Norway = No, Iceland = Is). In two other capitals (Denmark = Dk, Finland = Fi) victims could have a police-requested FME at a forensic institute, and be further

(13)

13 referred for treatment, but systematic socio-medical support was not established. Sweden had no single centre serving the capital area. Thus, totally four centres, each serving a defined population, were available for comparing the SAC organisation to the older model where only police could request FME from a forensic service.

The two SACs offered approximately the same services, the Is hospital-based and engaging gynaecologists on call, the No in a primary care emergency ward where the staff is basically trained in FME.

In the multicentre study, information from each centre was gathered by the author affiliated with the respective centre, according to a joint protocol.

Three centres provided data from 1996, Dk data from 1994 were extracted from a previous study (44).

The FMIs included all rape/sexual assault victims aged 12 yrs in one year. The SACs, operating with a lower age limit of 12/14 yrs (Is/No), included all cases except a couple of pre-school outliers referred to paediatric wards.

As population size is different in the four capitals (roughly Is 150,000, No 500,000, Fi 800,000 and Dk 1,000,000), results were presented in crude numbers, percentage within each centre, as well as “indexed” to a 100,000 female population at risk ((crude female

number/female populations aged 12 – 54 yrs) x 100,000). The latter allows direct

comparisons, being informative upon relation between victims/catchment populations most at risk. Focusing the population most at risk reduces bias due to inequalities in the prepubertal and the elderly population; and from adolescence on, females dominate as victims of sexual assault and visitors at SACs, while victims 55/60 yrs rarely attend (33;37;42;45-53).

In Papers II – IV the Oslo series were expanded by including the cases from 1999, as well as review of corresponding police files.

The study has been approved by the National Data Inspectorate, the Regional Research Ethics Committee and the Committee for Secrecy and Research (“Rådet for taushetsplikt og forskning”) for the judiciary system.

Annual reports from Oslo SAC have been rather stable since 2000 with regard to case profiles, the fractions forensically examined, and known reporting rates. The annual number of cases was approximately 150 during the centre’s first eleven years, increased in 1998/99 and stabilised at 205 ±15 thereafter. Although a second rise occurred in the second half of

(14)

2006, the years embraced by this study are considered representative and relevant to demonstrate points of interest.

Data regarding medical, forensic and counselling casework at Oslo SAC were collected from standardised SAC files. Variables concerning the assaults are based on victims’ descriptions (for details see Papers II– IV):

• Victim: gender, age, ethnicity, and additional vulnerability (i.e. physical/mental handicap, serious physical disease, diagnosed psychosis before/at/after consulting SAC, addiction problems, previous sexual abuse/assault)

• Interval between assault and presentation at SAC

• Type of sexual assault according to the most serious act: penetration of body orifice with penis/object; non-penetrative assault; amnesia/strong suspicion of assault; unclear cases/vague description

• Coercion: verbal, holding, violence in excess of holding, exploitation during alcohol/drug intoxication, and unclear/vague description. Assaults consisting of

several acts of coercion are coded according to the one most likely to result in bodily harm. Cases involving weapons are specified in text

• Number of perpetrators and victim’s relation to perpetrator: unknown, known other than partner, present/previous intimate partner

• Site of assault: victims’ area, perpetrator’s area, neutral. Geographically; venue within or outside Oslo precinct

• Forensic examinations: recorded extragenital and anogenital injuries. Trace evidence samples secured at FME

• Medical examinations performed (evidence of pregnancy, sexually transmitted disease) and attendance to medical follow-up (at least once)

• Counselling: whether victim attended one or more counselling sessions

• SAC documentation conveyed to the police: FME, medical/counselling/follow-up information, expert statements

Population data for calculating attendance rates for the female population most at risk were obtained from Statistics Norway (see www.ssb.no).

A population survey in SAC’s catchment area provided information on incidence of sexual assault among female Oslo inhabitants aged 24 – 55 yrs (54). This 2002 population incidence

(15)

15 was considered comparable to the 1999 incidence, thus allowing an estimate of the fraction of actually assaulted having attended SAC in 1999.

Some SAC files contained scant details of the assault due to victim’s inability/unwillingness to tell, incomplete examinations or inveterate cases. Lack of information resulting from the victim's inability to explain is specified since such cases occur regularly.

The lower age limit at this SAC is normally 14 yrs but during the study period, two 13 year- olds were included. One case of only counselling was included in the comparative study, but not in the following sections as the main file was missing.

In order to identify cases registered with the police, victims’ identities were cross-checked against the national population register. Several victims were found to have died and the year of death was noted; causes of death were not specified.

Police-registered cases were traced through national police registers (STRASAK, SANSAK).

Local Oslo files were searched for preliminary statements e.g. from police squads bringing victims to SAC. Thus, most probably all cases registered with Norwegian police were identified. Three more cases reported abroad were included for the evaluation of predictors of police involvement, but not when studying the police’s utilisation of SAC work.

For the original paper comparing SACs and FMIs, information on police registration in Oslo was collected from SAC’s internal records on police notifications and SAC documentation conveyed to the police. The subsequent access to police registers disclosed the true number of reporting women in Oslo to be 72, not 61, as the police had not contacted SAC in all reported cases. The Is SAC has not proceeded with similar research and thus not checked cases against police registers, but does not suspect distinct discrepancies as SAC cooperated closely with police and legal counsel. At the two FMIs, all cases were registered with the police. For this presentation the main results in the comparative section have been recalculated according to adjusted data from the Oslo SAC (Table 1).

Retrievable police files were reviewed. These comprised interrogations, technical reports, correspondence, and verdicts in cases brought to court. There were no written assessments except in the verdicts.

The police’s coding of a case was read from the STRASAK register at the time the search was performed (2005). Codes may be modified during investigation and thus diverge somewhat

(16)

from the code at first presentation of complaint, but not so that rape codes are eradicated from cases where investigation indicates no crime. The present study does not explore code modifications.

Six police files were inaccessible; core information from these cases was collected from STRASAK.

The following data were gathered from the review:

• Date of reporting, tertial of the year (see def)

• Intervals between assault, arrival at SAC and registration with the police

• Police classification of reports

• Victims’ withdrawal of consent to investigation

• Police identification of perpetrator(s)

• Police interrogation of perpetrator – whether perpetrator admitted main sexual acts in accordance with victim’s explanation. Other inconsistencies between victim/offender statements were not evaluated

• Cases where police requested analysis of trace evidence sampled at FME. Analysis results were coded according to the evidentiary strongest outcome (extrinsic DNA >

sperm > acid phosphatase). Trace evidence collected by police e.g. at site of assault was not included. Results were gathered directly from the laboratory

• Medical/supportive information and expert evaluations from other sources than SAC

• Legal outcome

Five female victims were seen after two separate assaults each, with different perpetrators. All these assaults are included as the study focuses on casework. For the same reason three police-reported cases linked by the same perpetrator were also included. More cases among those not reported, or with unidentified perpetrator, may have been similarly linked. Repeated FMEs involving same victim and same perpetrator e.g. in a violent relationship, were not encountered during these years.

Cases were separated according to victims’ gender for the comparative part of the study; but not for the following sections based on the files from Oslo SAC as numbers of males were few and gender differences were rare (Paper II).

(17)

17 Case patterns 1996 and 1999 in Oslo were quite similar and data were merged in Paper II IV.

Definitions

Victim/complainant: an individual alleging an incident of sexual violence against her/him Perpetrator: a person who, according to victim, has committed assault against victim Attending rate (females): number of cases per 100,000 population at risk. If not otherwise specified; limited to number of cases involving female victims aged 14 – 55 yrs related to corresponding selection of population at risk. The age-limits were set to cover the groups most at risk – 14 yrs being the lower age limit at this SAC, and sexual assaults are assumed less common after age 55 yrs.

Classic rape trilogy, the stereotype of “real rape”: penetrative assault by violence, unknown perpetrator

Forensic medical examination (FME): examination of the body surface and orifices for injury and trace evidence collection. Solely toxicological testing is not included as FME (few) Early cohort: arrival in time for FME. Time limits for FME with sampling changed between 1996 and 1999, from three to seven days post-assault (cervicovaginal swabbing). The cut-off for “in time” and early cohort was linked to the possibility of FME for the studies focusing FME

Late cohort: those arriving too late for FME. All but 6 hesitated > 7 days Reported cases: cases where the assault was registered with the police

Forensic benefit of self-referral: cases registered with police 2 days after FME at SAC. As loss of trace evidence mainly occurs during the first 24 – 36 hours post-assault; the possibility of finding evidence is markedly reduced at any FME performed later. Previous practice of performing FME only at police request often delayed examination towards or past this quality limit; or FME would not be requested at all. Victims consulting and reporting the same date

±1 day, would have stood a reasonable chance of early FME within a police-dependent organisation; but definitely not those postponing police involvement for two or more days.

Thus the forensic examinations performed in the latter cases represent the forensic benefit of self-referral SAC organisation.

Early withdrawal: victims’ consent to further police involvement withheld in the early phase of investigation. Most of these victims did not return to give a full statement after the first notification; and permission to collect medical evidence was not obtained

(18)

Late withdrawal: victims’ consent to cooperate withdrawn during ongoing investigation;

medical information had/ could have been collected

Rape, classified by police as rape: the definition according to Norwegian penal code is quite wide and includes insertion of penis/object/finger in vagina/anus or penis in mouth, as well as masturbation. Victims may be coerced by force, threats or during unconsciousness/drug- induced incapacitation

Other police codes: includes attempt at rape, sexual act with child <16 yrs, exploitation of dependency/profession, bodily harm/ threats, self-inflicted injuries as well as preliminary registrations by police not completed with a full victim statement

Tertial of complaint: which 4-months period of the year complaint was filed (1: Jan – April;

2: May – Aug; 3: Sept – Dec). The first two tertials were later merged as these results were similar.

Available cohort: cases where police had access to SAC case-work; victim permitting collection of SAC work, investigation not closed before victim’s arrival at SAC Forensic cohort: cases within available cohort where FME had been performed

Trace evidence subgroup: subgroup within forensic cohort where FME included collection of trace evidence

The non-forensic cohort: cases within available cohort where FME was not performed

Statistics

For the comparative part of the study, information from the four centres was collected in Excel. Significance of frequency differences were evaluated with chi-squared tests; z test in SPSS version for the original paper and 2008 Epi info when calculating the frequency differences related to population for this presentation. In the latter calculation the original figures were used, whereas Table 1 presents the results per 100,000 females.

In the sections concerning Oslo SAC, the statistical analyses were performed by use of SPSS version 11. Statistical significance of frequency differences were evaluated with chi-square test and Fisher’s exact test when any cell had an expected number 5. Unadjusted and adjusted odds ratios were calculated by means of logistic regression analyses.

Dependent variables in the logistic regression analyses were “presentation later than a week”

performed within total material, “registration with the police” within early and late presenting cohorts, “FME documentation utilised by the police” within forensic cohort and “trace evidence sample analyses performed” within the trace evidence subgroup.

(19)

19 The polytomous variables of assault characteristics were coded using the classic rape features as reference.

Age was linearly related to late arrival (increasing), showed a U-shaped relation to registration with the police, an S-shaped relation to analysing trace evidence (trend decreasing), and was unrelated to collection of FME documentation. Age had thus to be treated differently in the various analyses. For details, see Papers II, III and IV.

Univariate logistic regressions were first performed. Due to different numbers of cases in the various analyses, different approaches were chosen for the multiple analyses.

A backward regression was chosen for late presentation in the total material, taking into account that a valid result requires that the numbers of independent variables included at each step do not exceed one tenth of the smallest part of the dichotomously divided material. The most relevant univariate significant variables were entered into the first multivariate model, followed by stepwise exclusion of non-significant variables.

For analysing registration with the police in early and late cohorts, significant univariate variables were entered by a forward stepwise procedure.

When analysing the police’s use of FME documentation, only two variables were entered at a time, due to the low numbers of not collected records. The regression regarding forensic analyses were for similar reasons restricted to 6 variables. Restriction rested on clinical judgment, and we wanted to explore less obvious associations.

The final models were tested for goodness of fit. The statistical analyses were performed by use of SPSS version 11.

(20)

RESULTS

Paper I – summary

Different Nordic facilities for victims of sexual assault: a comparative study Nesvold H, Worm AM, Vala U, Agnarsdottir G.

Acta Obstet Gynecol Scand 2005;84:177 – 83

The first section encompassed 380 cases from four Nordic capitals; two FMIs (Dk, Fi) and two SACs (Is, No).

Core information concerning the 358 female cases is presented in Table 1 and Fig 1 in the erratum, where the Norwegian figures have been revised according to true incidence of police-registration. At both SACs, 53% of the cases were registered with the police; these cases are focused for the comparison.

Related to population at risk, SACs received 2 – 3 times more reporting victims than the FMIs, the differences mainly affecting those 16 – 24 yrs (Fig. 1), much less those older. There were clear differences between the SACs and the FMIs on a group level, but also some inter- centre differences within each group. The latter related mainly to attendance rates relative to population as the Is SAC and Fi FMI had 55% higher attendance rates than the No SAC and Dk FMI, respectively. In most aspects, the Is SAC and the Dk FMI were at each end of the scale, while the No SAC and the Fi FMI often were relatively close (Table 1).

The crude fractions showed that where the police regulated admission, mainly victims reporting 24 hours were sent for examination at FMI. Within the self-referral systems where victims were free to choose when to attend, arrivals were spread over a wide time span. Yet, relative to the size of the population, as many or more were seen acutely at SACs as at forensic institutes, and the SACs performed more FMEs than the FMIs.

Regarding the reported assaults the Dk FMI showed the highest crude fraction of penetrating assaults (rapes), Is SAC the lowest, i.e. Dk included few other assaults (attempts) whereas Is saw several kinds of assault. When controlling for population size, the results on penetrating assaults were reversed; Is SAC seeing fourfold more than Dk FMI. As for the attempts, the Is SAC showed considerably higher figures, both in crude fractions and relative to the

population; whereas the other three centres had pretty low and similar results (Table 1). The results on attempts constituted an important difference between the two SACs.

(21)

21 In the majority of cases at all centres, only one perpetrator was involved, and most

perpetrators were unknown or peripherally known to the victim (Table 1). However, when correcting for the size of the population, 5 – 7 times more assaults by known perpetrator were seen at SACs than referred to FMIs.

The post-hoc correction of numbers of police-reported in No resulted in only minor alterations regarding reported fractions of subgroups like multiple perpetrators/ violence including holding/ disclosed injuries at examination, as compared to the original publication where reporting was based upon information from SAC alone (corrected data not shown¹). In both the original and the revised version, the Is SAC diverged markedly from the other centres by higher attending rates, a higher number of attempted rapes, less use of violence and weapons and less injuries in addition to the victims being younger.

Paper II – summary

Sexual assault centres: Attendance rates, and differences between early and late presenting cases

Nesvold H, Friis S, Ormstad K.

Acta Obstet Gynecol Scand 2008;87:707 – 15

A total of 354 cases presented at Oslo SAC during the years 1996 and 1999; 6% males, 8% of non-Western origin.

Attendance rates showed that 0.12% of the female at-risk-population aged 14 – 55 yrs consulted SAC in 1999; 0.31% in the 14 – 23 age group and 0.07% in the 24 – 55 age group.

In the older age group, the Oslo population survey disclosed a 1 – 2% annual incidence of sexual violence (1% when focusing rape, 2% when including rape, attempt at rape and forced sex) (54). Thus, in this age group, an estimated 3.5 – 7% of all female victims of sexual assault presented at SAC. There was no survey suitable for comparison with the younger age group.

Cases presenting in time for FME (early cohort n=278, 78.5%) differed from those later presenting (late cohort, n= 76, 21.5%) in several aspects.

1 In general, all previous results regarding reported cases are maintained or strengthened, none reversed.

(22)

Early arriving victims were older (mean age 28.3 yrs) and included 90% of those with an addiction problem. A majority of 78% was escorted to SAC, 31% by police. The perpetrator was a stranger in 55% of the cases. Two thirds of the early presenting assaults were penetrative; the remaining third included almost 90% of all non-penetrative and suspected/vague cases. Forty-eight (17%) of early presenting victims were hospitalised, usually for a short observation in the emergency ward, and 55% of the medically examined complied with follow-up.

The late cohort of 76 cases comprised more adolescent and young victims (mean age 24.3 yrs); almost half of these were escorted. There were less non-acquainted perpetrators (36%), the sexual acts mainly penetrative (85%). Methods of coercion among late arrivals did not deviate significantly from the early cohort, except that verbal coercion significantly increased the odds for late arrival. Victims in the late cohort were medically examined in 35 (46%) of the cases, and 80% of these complied with follow-up (n.s.). Only one person was hospitalised.

Adjusted odds demonstrated an association between young age and late arrival, as odds decreased by 0.96 (CI 0.93 – 0.99) per increasing year. Odds for late arrival were increased if the perpetrator was a partner (3.6, CI 1.4 – 10.0) or an acquaintance (2.5, CI 1.4 – 4.4), as compared to strangers. Odds were also increased where verbal coercion was present (2.9, CI 1.1 – 7.7) as compared to violence exceeding holding. Merely suspected abuse and cases of vague explanation showed reduced odds (0.3, CI 0.1 – 0.7) compared to the penetrative assaults.

Counselling (300 cases in total) and further referrals occurred equally often in both cohorts;

44 (12%) somatic and 138 (39%) psychiatric. Several cases affected particularly vulnerable victims, totally 163 (46%) cases, 158 victims. The five females attending twice, all in the early cohort, had all been exposed to sexual abuse prior to first consultation, and two had an addiction problem. Thus total numbers of previously abused victims are 109, total numbers of addicts 59. The 21 physically/mentally handicapped and the 13 diagnosed with psychosis attended only once each. Apart from victims with an addiction problem, vulnerable groups seemed just as often present among early and late presenting victims.

Nineteen (5%) of the victims were deceased by 2005, their median age at consultation was 30 yrs (range 20 – 59 when excepting two > 80 yrs).

(23)

23 Paper III – summary

Sexual assault centres and police-reporting - an important arena for medical/legal interaction

Nesvold H, Ormstad K, Friis S.

Ref Type: Submitted to Journal of Forensic Sciences

Of the 354 cases seen at SAC, 180 cases were registered with the police, 177 with Norwegian police.

In the early cohort (n=278) where victims attended in time for FME, 154 (55%) were reported; while 26 (34%) of the cases were reported among victims presenting later than a week post-assault. Median interval from assault to reporting was one day in the early cohort and 33 days in the late cohort.

In the early cohort, the following variables were significantly related to increased likelihood of reporting: age 30 yrs, serious handicap, violence more than holding, single perpetrator, police escort to SAC, the use of SAC services, presence of physical injuries. Variables related to reduced likelihood were: age 18 – 29 yrs, drug/alcohol addiction, exploitation during inebriation, suspicion of involuntary intake of alcohol/drugs; vague description of or inability to describe sexual act/coercion/perpetrator; as well as absence of physical injuries at examination.

In the late cohort, serious handicap was significantly related to increased likelihood of reporting, while individuals who had previously been subjected to sexual abuse were less likely to report.

FME was performed in 238 cases, 142 (60%) of these were registered with the police.

Consequently, 96 (40%) of the FMEs at SAC were performed “in vain” since these assaults were not registered; however, 17 of these victims were escorted by police and police- involvement was taken for granted.

In the early cohort, 111 (70%) of the reporting victims consulted SAC and police within same day ±1. In 34 cases, FME preceded reporting by 2 days. In these cases FMEs would have lost considerable quality if FME had to await a police request, as occurs in police-dependent organisations. Consequently, these 34 cases represented the police’s forensic benefit of self- referral, constituting 24% of the reported FMEs.

(24)

In the late cohort, police contact occurred equally often prior to SAC as after, 57% of the cases within ±1 month of SAC consultation, total range from four years before to two years after, and with no definite peak related to consultation.

We made three logistic regression analyses to see which variables most strongly predicted the likelihood of reporting.

The first two analyses comprised the early cohort only. The first one restricted the independent variables to victim and assault characteristics, i.e. information available before examination. Age 30 yrs was related to increased likelihood (OR 3.1, CI 1.7 – 5.8), whereas reduced odds were seen if victim had an addiction problem (OR 0.4, CI 0.2 – 0.7), if victim was exploited during intoxication (OR 0.4, CI 0.2 – 0.9) or was unable to describe coercion (OR 0.0, CI 0.0 – 0.5).

The second analysis added use of services/examination results as independent variables. The most important variables related to increased likelihood were age 30 yrs (OR 3.2, CI 1.7 – 6.1) and age <18 yrs (OR 2.7, CI 1.1 – 6.9), injuries disclosed at FME (OR 2.9, CI 1.1 – 7.4, where only extragenital injuries were found; OR 20.1, CI 4.3 – 94.9, when both anogenital and extragenital injuries were present), and compliance with medical follow-up (OR 2.8, CI 1.6 – 5.0). Inability to describe perpetrators/no information was associated with reduced odds (OR 0.05, CI 0.01 – 0.3).

The third analysis comprised the late-comers only and identified no variables of significant influence.

The attrition, of cases from SAC to police related partly to victims’ behaviour and partly to police decisions.

In a total of 174 (49%) of the cases seen at SAC, the victims avoided police registration; 29 of these had been police-escorted to SAC without returning to file a complaint. Among the police-registered cases 24 victims withdrew their complaints; 11 early, 13 during ongoing investigation; all these were in the early cohort.

Regarding the police-related attrition; according to recommendations, all 91 police-escorted victims should have been registered at least with a preliminary report, but only 24

preliminaries were seen. The 29 police-escorted victims not registered were thus among the 67 without any preliminary and consequently barred from outreach by police; these 29 did not readdress at own initiative.

(25)

25 The police might have referred more victims to SAC. Totally 197 cases of rape were registered in the Oslo precinct these years, 103 of these presented at SAC, 94 (47%) did not.

Among our police-registered victims, we identified seven reporting swiftly in time for FME but attending too late for proper FME, and four arriving after the investigation had been closed.

Paper IV – summary

To be used or not to be, that is the question.

Legal use of forensic and clinical information collected in a self-referral Sexual assault centre

Nesvold H, Ormstad K, Friis S.

Ref Type: Submitted to Journal of Forensic Sciences

Among the 177 SAC cases registered with the Norwegian police. SAC information remained accessible to the police in 163 cases. In 134 of these, FME-based data were available (forensic cohort); and in 118 cases the FME included trace evidence sampling (trace evidence

subgroup). Only clinical information was available in 29 cases, in which forensic examinations had not been performed (non-forensic cohort).

The police did not collect all available SAC information. FME documentation was requested in 112 (84%) cases and 60 (51%) of the trace evidence kitswere analysed. In the non-forensic cohort SAC-documentation was collected in only 9 cases (31%).

Request for FME documentation was clearly associated with cases police-classified as rape and complaints filed during the first eight months of a year (OR 11.1, CI 3.5 – 34.9 and OR 4.2, CI 1.4 – 12.5). Victim being drug/alcohol-addicted was associated with reduced odds for collecting documentation (OR 0.3 CI 0.1 – 0.9).

Similar associations were found regarding trace evidence analysis; cases police-classified as rape and complaints filed during the first eight months of the year showed increased odds (OR 6.3, CI 1.4 – 28.5 and OR 6.7, CI 2.4 – 18.3), similarly if victims were < 20 yrs (OR 6.9, CI 2.1 – 22.8). Addiction was unrelated, whilst cases where the site of assault was perpetrator’s area showed reduced odds (OR 0.2, CI 0.1 – 0.5).

(26)

Interaction analyses indicated that collection of documentation by the police was even more determined by police-classification, time of the year and non-addiction when perpetrator was unidentified; the high odds were 5 – 10 folds higher and the low odds similarly lower than with a known perpetrator. However, several confidence intervals were wide and the results are therefore somewhat uncertain. The results are referred as they have clinical relevance.

In the non-forensic cohort, information was only collected in cases police-classified as rape and all but one perpetrator was known to the victims.

As documented above, the likelihood that the police would use the SAC information, was clearly related to whether the police classified a case as rape or not. It is therefore interesting that the police classification was only partly consistent with victims’ description at SAC. In 11/133 (8%) cases described as penetrative or strongly suspected penetrative assault; the police used other codes than rape (e.g. sexual act with child, exploitation of dependency; see Paper IV, Table I). Oppositely, 9/30 of those described as non-penetrative or vague at SAC were still classified as rape by the police.

Practical use of SAC casework

Trace evidence analyses were requested in 60 cases; extrinsic DNA detected in 27 cases, 21 results matching a suspect. Results supporting sexual contact (anogenital injuries and/or sperm/extrinsic DNA) were found in 68 cases. Where such evidence was present, a higher proportion of perpetrators admitted sexual contact 34/43 (79%) vs. 28/46 (61%) with no such evidence (p=0.06). Sexual contact was denied in 27 cases; police could refute the claim in 9, but did not explore these opportunities in 15 cases² as analyses were not requested.

Among 79 initially unidentified perpetrators, 41 were eventually identified, 8 (20%) identifications confirmed by DNA.

Most FME files documenting injuries were collected, including the 37 most severe ones, but only in one case expert interpretation of injuries was requested.

Supportive non-forensic evidence like documentation of post-assault consequences was present in 45% of the cases irrespective of cohort, mainly conveyed as short notes at SAC’s initiative. Information that police had to request specifically, e.g. from counselling or family

2 In three cases analyses had been performed, but no seminal constituents found.

(27)

27 physician, was collected in 24% of available cases. However, elaborate descriptions and interpretation of psychosocial sequelae were not present in the police files, and had according to SAC not been asked for.

SAC information and legal outcome

Table 2 includes all cases in these series registered with the Norwegian police, and is assembled especially for this presentation to demonstrate the overall result of attrition. The table presents key information from Paper II – IV, sorted according to FME-based documentation being present in the police files or not. Legal outcome is included, as the impact of FME and SAC information is an issue of concern.

The cohort where FME-documentation was present showed higher proportions of cases police-classified as rape, early presented complaints, forensic examinations of crime scene, and files including documentation of sequelae. The latter three proportions were even higher among cases taken to court (data not shown).

Significantly more cases were taken to court and fewer cases were classified as non-criminal where FME documentation was present.

(28)

(29)

Table 1. Paper I. Inter-centre comparison: revised presentation of case profiles

Data on 358 female cases of sexual assault, seen at Forensic Medical Institutes (FMI)/Sexual Assault Centres (SAC) in Denmark (Dk, FMI), Finland (Fi, FMI), Norway (No, SAC) and Iceland (Is, SAC) during the study period.

Figures in bold represent police reported cases per 100,000 females at risk (14 – 54 yrs). Exact figures are given in parentheses; percentages are calculated within total number cases with the respective information at each centre.

The presented Norwegian results are based on corrected identification of police-registered cases.

Regarding relationship between complainant and perpetrator, this version includes cases with both single and multiple perpetrators (among the latter, all unknown/peripherally known), but not cases with no information.

Police reported Not reported

Dk n=62

Fi n=86

No n= 72

Is n=39

No n= 64

Is n=35 Mean age in years

[range]

24 [12-51]

28 [14-47]

28 [15-59]

24 [12-48]

26 [14-80]

25 [14-58]

Number/100,000 women at risk †

Age group 12-54

18 ** 28 ** 48 ** 78 42 * 68

Performed FME per 100,000 at risk

(n)

15 ***

(53)

28 0.07‡

(85)

38 ns (57)

56 (28)

25 ns (37)

36 (18) Arrived within 24 hrs

per 100,000 at risk;

(n);

%

14 ***

(50) 81%

27 ns (81)

94%

23 * (35) 49%

42 (21) 54%

15*

(23/63 ) 37%

32 (16) 46%

Arriving > 7 days; % (n)

13%

(8)

0 (0)

15%

(11)

28%

(11)

24%

(15)

31%

(11) Rape (penile penetration)

(n);

%

16 * (55/ 61)

90%

22 **

(67/86) 78 %

35 ns (53/72)

74%

50 (25/39)

64%

30 ns (45/63 )

71%

46 (23/35)

66%

Attempted rape per 100,000 at risk;

(n);

%

1 ns (4/61)

7%

0,7 **

(2/86) 2%

4**

(6/72) 8%

18 (9/39)

23%

3 ns (4/63 )

6%

4 (2/35)

6%

Perpetrator unknown / peripherally known per

100,000 at risk;

(n);

%

12 **

(44/57) 77%

21 ns (63/79)

80%

28 ***

(42/72) 58%

56 (28/39)

72%

21 ***

(31/55) 56%

54 ( 27/35)

77%

Perpetrator known , including intimate partner

(n);

%

3 ns (13/57)

23%

4 ***

(16/79) 20%

20 ns (30/72)

42%

22 (11/39)

28%

16 ns (24/55)

44%

16 (8/35)

23%

† When calculating attending rates per 100,000 females aged 12 – 54, three cases affecting older women (1 Is, 2 No) were excluded in the calculation of attending rates. These had to be included when calculating other figures per population at risk, as the common database was country-wise, not individual.

* p 0.05 ** p 0.01 *** p 0.001

Chi squared tests were calculated for 2x2 tables entering the exact fractions from two neighbouring columns; the labelled one with the one to the right. Thus the p-value results presented for Dk refers to the comparison between Dk and Fi, the Fi – No comparison is presented with Fi, and the No-Is comparison with No.

‡ Regarding FME; the p value was 0.001 for the Fi / Is fractions.

(30)

Table 2. Paper IV. Presence of FME documentation in police files and legal outcome

The 177 cases registered with the Norwegian police, divided in Cohort I where FME documentation had been requested by police, and Cohort II where no FME documentation was present in the police files. Various forms of attrition caused the abscence; FME had not been performed, was not available or documentation had not been requested. FME was not performed if victim was unwilling or arrived too late.

Legal outcome and important background information of the cases in each cohort is presented

COHORT I COHORT II TOTAL FME documentation

present in police files

FME documentation

not present

n = 112 % 65 % 177 %

LEGAL OUTCOME

To court, total * 27 24,1* 5 7,7 32 18,1

- Convicted penalty code 24 3 29

- Civil conviction 1 1 1

- Acquitted 2 0 2

- False complaint, convicted 0 1 1

Fined 1 0,9 2 3,2 3 1,7

Dismissed 76 67,9 42 61,7 118 66,7

- Evidentiary reasons total 49 43,8 22 33,8 71 40,1

- Perpetrator not identified 25 22,3 16 21,3 41 23,2

- Other reasons 2 1,8 4 6,6 6 3,4

No- crimed *** 8 7,1*** 16 24,6 24 13,6

BACKGROUND

Cases with identified perpetrator ¤ 84 75,0¤ 40 61,5 124 70,1

Police-classified as rape 98 87,5 41 61,5 139 78,5

Victim’s description at SAC:

Penetrative assault 84 75 42 64,6 126 71,2

Non-penetrative assault 13 11,6 16 24,6 29 16,4

Amnesia, suspected assault 12 10,7 6 9,2 18 10,2

Vague story 3 2,7 1 1,5 4 2,3

Interval assault – police

<4days 87 77,7 32 49,2 119 67,2

> 7days 13 11,6 30 46,2 43 24,3

No of scene investigations 38 33,9 7 11,5 46 26,0

No of supplementary doc 60 53,6 14 21,5 74 41,8

Withdrawals of complaint

- early 0 11 _16,9 11 6,2

- late 11 9,8 2 13 7.3

(31)

Comments to Table 2

In the 65 cases where no FME information was present in the police files;

- in 14 cases information was unavailable to police due to early withdrawal of complaint or investigation was closed before arrival

- in 22 cases FMEs were available

- in 59 cases only other SAC information was available

As more perpetrators were identified in the cohort where FME documentation was present, each cohort was subgrouped according to identified perpetrator or not; data not shown. The main differences between the subgroups related to numbers of cases taken to court and dismissals due to unknown perpetrator.

Among cases taken to court:

In cohort I, 27 cases were taken to court; 21 (78%) were classified as rape by the police; 24 (90%) were reported

<4 days; 16 (59%) included a site of assault examination/brief inspection; 18 included supplementary medical documentation.

In Cohort II, five cases were taken to court; 4 were rape-classified; 2 were reported within 4days; 2 included a brief inspection at site of assault; 3 included supplementary documentations.

Dismissals:

Fractions of dismissals due to evidentiary factors were similar in the two cohorts.

In cohort I, all but two cases of unidentified perpetrator were “correctly” dismissed due to unknown perpetrator.

In cohort II, only two thirds of the cases of unidentified perpetrator were thus dismissed and 8/25 were no-crimed.

No-criming occurred more commonly in Cohort II; in 7 of these cases victims had withdrawn the complaint early.

In cohort I, 3 late withdrawals were no-crimed.

(32)

(33)

DISCUSSION

In cases of sexual assault, medicolegal examinations at police request have been performed at forensic medical institutes (FMI), and within wards of gynaecology and emergency. Self- referral centres for victims of rape and sexual assault (SAC) have been established since the seventies.

Previous reports from these services have focused on victims and assaults; examination results related to victims’ age (55), to described violence or to the relation between victim and perpetrator (35;52;56); anogenital injuries related to forced and voluntary sexual acts (57-59);

factors associated with follow-up (60), police-involvement (30;42;43;61) and legal outcome (8). This way, medical literature has strived to improve the fundaments of expert statements, hoping to improve legal outcome.

This thesis explores inter-centre differences in attending patterns (Paper I), and the Oslo SAC’s success in reaching the target group, how SAC services are used by victims and legal authorities, police-involvement, medical/legal interactions, the attrition and the effect on legal outcome (Paper II –IV).

In the following, head issues from each paper are discussed separately before the overall discussion.

Paper I

Comparison of different services for sexually assaulted

Self-referral sexual assault centres (SAC) and forensic medical institutes (FMI) are principally different facilities which are both providing medico-legal documentation. SACs serve two intentions: to assist victims and to provide medico-legal information, whilst FMIs serve only the latter.While victims may consult SACs directly for any help they may need, the police regulate admission at FMIs, and cases have passed three levels of selection before

presentation; victims have to approach the police within time for FME, the police must decide whether to request an examination, and victims must accept the examination.

(34)

Thus, terms for attending differ between SACs and FMIs. In a comparison, the impact should be searched in the patterns/profiles of police-reported cases as FMIs see only these. At SACs, assistance to non-reporting victims represents an additional gain.

The results of Paper I clearly show major differences between SACs and FMIs concerning number of cases in relation to catchment population (attending rates), interval to arrival, case profiles including age and victim/perpetrator relationship. When related to population, both SACs receive more assault victims, especially among adolescents/young adults and those coerced by a known perpetrator, as well as those hesitating to attend. Attending rates were however rather similar for those > 24 yrs regardless of type of centre. The SACs also receive more acute-presenting victims and perform more forensic medical examinations than the FMIs.

The recalculated figures per 100,000 females at risk provide the best fundament for inter- centre comparisons, disclosing divergences between the two kinds of organisations, as well as variations among services of similar organisation. In most aspects we find a falling gradient;

the Is SAC seeing most, followed by No SAC, Fi FMI and Dk FMI.

Several differences seem robust. Some are clearly linked to attending rates, like performed FMEs; whereas others, like age and victim/perpetrator relation, illustrate that the services attract different cases.

The observed differences may result from discrepancies in occurrence of sexual assaults and/or in willingness and barriers to seek help/report. At FMIs the police are the final gatekeeper, and their attitudes towards requesting examination are decisive. So far, information on these topics is limited.

Incidence vs. thresholds

Population surveys on incidences vary in methodology and are difficult to compare. Most Nordic surveys focus on domestic violence (54;62-64), some describe total incidence of sexual assault (54;62;65).

Yet, last years’ incidence of sexual assault among Dk women 18 – 70 yrs is lower than estimates from many other North European/Western countries, being 0.2% (65) vs. 0.5 – 2%

(7;28;54;62;63;66;67)³. In a Nordic multicentre study, Dk shows the lowest lifetime

3

HEALTH SERVICES FOR SEXUALLY ASSAULTED. COMPLAINANTS’ AND JUDICIAL USE OF A SELF-REFERRAL CENTRE; CASE-FLOW FROM ASSAULT TO LEGAL OUTCOME